Model comparison
Gemma 3 27B vs GPT-4.1 nano
Gemma 3 27B is the stronger model overall, scoring 30.8 to 27.9 on the Noometry Index.
Last verified . 28 shared benchmarks.
GoogleSummary
- They share 28 benchmarks with published results for both. Gemma 3 27B scores higher in 7 categories and GPT-4.1 nano in 3 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 40.5.
- The biggest single-benchmark swing is Fiction.LiveBench: 33.3% for Gemma 3 27B and 25% for GPT-4.1 nano.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.10 / $0.40 for GPT-4.1 nano.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 27B | GPT-4.1 nano | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 30.8 | 27.9 |
| Released | 2025-03-11 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.05M |
| Max output | 8K | 33K |
| Input $ / M tokens | $0.08 | $0.10 |
| Output $ / M tokens | $0.16 | $0.40 |
| Results tracked | 43 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.1 nano leads
Gemma 3 27B: 22.5 (#334), GPT-4.1 nano: 24.1 (#330)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| Aider Polyglot | 4.9% | 8.9% |
| SciCode | 21.2% | 25.9% |
| LMArena Coding | 1322 | 1306 |
| WeirdML | — | 19% |
| LiveBench Coding | 39.9% | — |
Agentic & Tool Use GPT-4.1 nano leads
Gemma 3 27B: 25.1 (#110), GPT-4.1 nano: 26.5 (#104)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 33% |
Reasoning Gemma 3 27B leads
Gemma 3 27B: 16.7 (#301), GPT-4.1 nano: 8.5 (#349)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 33.3% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1340 | 1286 |
| DTBench | 52.5% | 52.5% |
| LMCA | 12.3% | 5.5% |
| Epoch Capabilities Index | 130.04 | 129.62 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| LiveBench | 50% | — |
Math GPT-4.1 nano leads
Gemma 3 27B: 25.9 (#265), GPT-4.1 nano: 26.9 (#252)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 28.9% |
| LMArena Math | 1312 | 1274 |
| MATH Level 5 | 74% | 70% |
| Omni-MATH | — | 36.7% |
| LiveBench Math | 55.4% | — |
| FrontierMath (Feb 2025 set) | — | 1% |
Knowledge Gemma 3 27B leads
Gemma 3 27B: 25.5 (#261), GPT-4.1 nano: 21.8 (#273)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| GPQA Diamond | 47.7% | 48.9% |
| LMArena Expert | 1304 | 1272 |
| SimpleQA Verified | — | 6% |
| MMLU-Pro | — | 55% |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 50.7% |
Multimodal Gemma 3 27B leads
Gemma 3 27B: 32.6 (#100), GPT-4.1 nano: 29.2 (#113)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| LMArena Vision | 1164 | 1063 |
| GeoBench | 52% | — |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 46.9 (#155), GPT-4.1 nano: 41.6 (#205)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| LMArena Non-English | 1334 | 1260 |
| LMArena Chinese | 1346 | 1270 |
| LMArena German | 1362 | 1288 |
| LMArena Japanese | 1287 | 1198 |
| LMArena Russian | 1349 | 1261 |
| LMArena French | 1368 | — |
| LMArena Korean | 1308 | — |
| LMArena Spanish | 1349 | — |
Instruction Following Gemma 3 27B leads
Gemma 3 27B: 70.6 (#160), GPT-4.1 nano: 67.8 (#193)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| LMArena Instruction Following | 1321 | 1267 |
| LiveBench Instruction Following | 74.9% | — |
| IFEval | — | 84.3% |
Long Context Gemma 3 27B leads
Gemma 3 27B: 27.6 (#293), GPT-4.1 nano: 23.7 (#296)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| Fiction.LiveBench | 33.3% | 25% |
| LMArena Longer Query | 1333 | 1283 |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.5 (#168), GPT-4.1 nano: 40.5 (#243)
| Benchmark | Gemma 3 27B | GPT-4.1 nano |
|---|---|---|
| LMArena Text | 1358 | 1285 |
| LMArena Creative Writing | 1346 | 1260 |
| EQ-Bench Creative Writing | 1266 | 946 |
| LMArena Multi-Turn | 1345 | 1277 |
| Short-Story Creative Writing | 79.9% | — |
| WildBench | — | 81.2% |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than GPT-4.1 nano?
Gemma 3 27B is the stronger model overall, scoring 30.8 to 27.9 on the Noometry Index.
Which is cheaper, Gemma 3 27B or GPT-4.1 nano?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; GPT-4.1 nano lists at $0.10 and $0.40.
Is Gemma 3 27B or GPT-4.1 nano better for coding?
GPT-4.1 nano scores higher on coding benchmarks: 24.1 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 nano does, with 1.05M tokens against 131K.
How many benchmarks do Gemma 3 27B and GPT-4.1 nano share?
28 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and GPT-4.1 nano has 38.